run_payment_operation

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does substantial work: it discloses that payee/currency/destination are hard-bound by the operator, that approval is required above stated amounts, that note fields reject URLs/emails/bank identifiers, and that amounts are range-bounded. It stops short of stating irreversibility, idempotency/retry semantics, or what a caller should expect on failure — material omissions for money movement.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the single most important constraint (what the agent may and may not set), then a scannable per-operation list. Dense and mostly waste-free, though the inline JSON schemas duplicate schema content and lengthen the block considerably.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

A polymorphic dispatcher with nested params and no output schema demands more than this provides: return values are never described (no output schema to fall back on), the 'reference' field mapping is absent, and error/ambiguity behavior across six operations is unstated. The operation catalogue is thorough, but the call-and-response contract is only half covered.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

At 67% schema coverage the baseline is 3, and the description genuinely compensates for much of the gap by inlining per-operation sub-schemas with patterns, enums and max lengths. The critical weakness is that the top-level 'reference' parameter is never named: the text says an operation 'requires a order_id this broker issued' without telling the agent to place it in 'reference', leaving a mapping the agent must guess.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The opening sentence states a specific verb and resource (run a payment operation) and immediately bounds the scope: only the operation name and its parameters are agent-chosen, while payee, currency and destination are operator-fixed. The six enumerated operations make the tool's surface unambiguous. It does not, however, name the sibling list_payment_operations or explain how the two differ.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

Each operation entry carries its own preconditions ('requires a order_id this broker issued', 'requires a capture_id this broker issued') and approval thresholds (over_amount:100.00, over_amount:200.00), which tells the agent when a call is viable and when escalation kicks in. What is missing is any explicit routing guidance versus list_payment_operations, or advice on which operation to pick in a given situation.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.